There have been many efforts to assess the intensity, frequency and duration of PA among males and females and people of all ages, but few efforts to objectively assess population level measures of PA. Determining population level PA by assessing at the individual level may neither the most appropriate nor cost- effective methods. Population level measures of PA will be important when assessing the effectiveness of community level interventions or in understanding how changes in the built environment might affect PA on a broad scale. We have already developed a tool to assess population level PA in parks with SOPARC (Systematic Observations of Play and Recreation in Communities) and tested its utility and reliability in Los Angeles. This tool allows the documentation of the number of park users, their age group, gender, race/ethnicity and activity levels. Our specific aims are: 1) to validate SOPARC with respect to its ability to measure park use and population level energy expenditure. 2) To test the reliability and validity of self report of physical activity in public parks and 3) To assess SOPARC in different populations, geographic settings and seasons in order to determine its utility on a national basis, 4) To develop a SOPARC manual and software package for dissemination to other researchers and key stakeholders interested in physical activity promotion and measurement. This proposal will include validation of observations using videography, expanded use of SOPARC in 10 parks across the US (Los Angeles (West Coast). Philadelphia (East Coast), Chapel Hill, NC (South), Columbus, Ohio (Midwest) and Albuquerque, New Mexico (Southwest), assessing physical activity on an hourly basis, 14 hours per day during daylight over a 2 week period;following 48 individuals in each site (n=240) using GPS monitors and accelerometers to validate their self-report of use of parks, including frequency of park visits, duration of visit, and activity levels in local parks over a 3-week period;and the use of SOPARC in 6 neighborhood parks in the 4 areas not previously studied in each of three seasons, spring, summer and fall. The US has an extensive infrastructure of parks that serve an estimated 75% of the population. In Los Angeles, parks were cited as the place where most people engage in leisure exercise. Therefore, parks may be the ideal venue for increasing PA among Americans. If SOPARC and/or self-report of park use can be validated, and the tools useful in different geographies, SOPARC can be used for assessing future community level interventions to promote PA as well as for population level surveillance of Our study will rigorously test and validate the method, System for Observing Play and Recreation in Communities (SOPARC), for measuring population level physical activity in neighborhood parks. The US has an extensive infrastructure of parks that serve an estimated 75% of the population and may be the ideal venue for increasing PA among Americans. SOPARC can potentially be used for assessing future community level interventions as well as for population level surveillance of PA.